Kelvin Fei Chu

Q10 or Q10dB value

2008-09-16T13:08:00.002-06:00

Q10dB value = center frequency / bandwidth of the tuning curve 10 dB above tip(threshold)

Auditory Physiology Basics

2008-09-16T11:53:00.002-06:00

What is a vocoder

2008-07-08T20:49:00.000-06:00

Install Sopcast on Ubuntu 7.10

2008-03-20T17:53:00.001-06:00

Sopcast really rocks!

How to make Ubuntu 7.10 Chinese font look right with Yahei font

2008-03-20T15:04:00.004-06:00

http://www.osxcn.com/ubuntu/edgy-stheiti.html
http://www.osxcn.com/ubuntu/ubuntu-stheiti.html
http://www.osxcn.com/ubuntu/ubuntu-stheiti2.html

These are some additional notes:

1. Create language-selector.conf in /etc/fonts/

$ sudo fontconfig-voodoo -f -s zh_CN

2. Download STHeiti.tff from

http://www.isload.com.cn/store/8m385v6af6trv/%D1%C5%BA%DA%CF%B8%BA%DA%20STHeiti.ttf/downlaod

Acer 5315 Sound Card Setup in Ubuntu

2008-03-19T22:32:00.001-06:00

Ubuntu on Acer Aspire 5315-052G12Mi

2008-03-19T22:31:00.001-06:00

Acer 5315 Wifi (Atheros AR5006EG/AR5007eg) driver on Ubuntu

2008-03-19T22:23:00.001-06:00

Ubuntu is highly suggested. Fedora 8 may meet with some APIC fatal errors.

Fedora 8 installation stops at: running/sbin/loader on laptop

2008-03-18T19:50:00.000-06:00

Solution:

http://forums.fedoraforum.org/forum/showthread.php?t=171230&highlight=%2Fsbin%2Floader

Power and Energy

2008-02-26T10:55:00.005-06:00

For a digital signal x(n) the power = mean(x.^2) and energy = sum(x.^2).

Importantly, Parseval's theorem is often written as:

Alternatively, for the discrete form, the relation becomes:

where X[k] is the DFT of x[n], both of length N.

A demo script is as follows:

Download: here.

Please give reference if you want to use the code.

Forward-Backward Filtering

2008-02-25T09:33:00.011-06:00

A very good illustration for how to cancel phase delay of filters by using forward and backward filtering.

http://ccrma.stanford.edu/~jos/filters/Forward_Backward_Filtering.html

Note:
1. If the filter were complex, then we would need to conjugate its coefficients when running it backwards.

2. Before applying the backward filtering, zero padding may be needed to avoid aliasing, as you may see from the implementation below. (take out the corresponding zero padding line)

The following is a implementation that cancels the delay of FIR bandpass filter banks.
Download: here.

Please give reference if you want to use the code.

Just noticeable difference or jnd

2008-02-15T09:47:00.000-06:00

From Wikipedia:

In psychophysics, a just noticeable difference, customarily abbreviated with lowercase letters as jnd, is the smallest difference in a specified modality of sensory input that is detectable by a human being. It is also known as the difference limen or the differential threshold.

Sensitivity index or d'

2008-02-15T09:44:00.001-06:00

From Wikipedia:

The sensitivity index or d' (pronounced 'dee-prime') is a statistic used in signal detection theory.

d' measures the separation between the means of the signal and noise distributions in units of the standard deviation of the noise distribution. An estimate of d' can be found from measurements of the hit rate and false-alarm rate.

A higher d' indicates that the signal can be more readily detected.

ROC curve (Receiver operating characteristic)

2008-02-15T09:39:00.001-06:00

From Wikipedia:

In signal detection theory, a receiver operating characteristic (ROC), or simply ROC curve, is a graphical plot of the sensitivity vs. (1 - specificity) for a binary classifier system as its discrimination threshold is varied. The ROC can also be represented equivalently by plotting the fraction of true positives (TPR = true positive rate) vs. the fraction of false positives (FPR = false positive rate). Also known as a Relative Operating Characteristic curve, because it is a comparison of two operating characteristics (TPR & FPR) as the criterion changes

Properties of Fourier Transform

2008-02-12T16:22:00.001-06:00

http://en.wikipedia.org/wiki/Fourier_transform
http://mathworld.wolfram.com/FourierTransform.html

Compute pseudo-frequency for wavelet scalogram

2008-02-12T14:13:00.000-06:00

% Compute pseudo-frequencies for wavelet scalogram
f = centfrq(wfun)./(a.*delta);

a is the scaler
wfun is the mother wavelet function
centfrq is the fft peak frequency
delta is sampling period (i.e. 1/fs)

Envelope signal after hilbert transform has a group delay of N/2

2008-02-12T13:55:00.000-06:00

Suppose the hilbert transform involves a N point FFT, then the group delay of the envelop signal is N/2.

Frequency -> ERB_N-rate and ERB_N-Bandwidth (Glasberg and Moore, 1990)

2008-02-11T01:04:00.000-06:00

ERBrate = 21.4.*log10(4.37*cf/1000+1);
ERBwidth = 24.7.*(4.37*cf/1000 + 1);

An Efficient Implementation of Gammatone Filters by Cooke

2008-02-08T10:48:00.001-06:00

http://www.dcs.shef.ac.uk/~ning/resources/gammatone/

Also: see http://library.wolfram.com/infocenter/MathSource/2052/ by Slaney

Doc:http://citeseer.comp.nus.edu.sg/cache/papers/cs/4963/http:zSzzSzwww.slaney.orgzSzmalcolmzSzapplezSztr35zSzPattersonsEar.pdf/slaney93efficient.pdf

Equivalent Rectangular Bandwidth for an arbitrary band-pass filter

2008-02-08T09:52:00.009-06:00

he equivalent rectangular bandwidth (ERB) of a filter is defined as the width of a rectangular filter whose height equals the peak gain of the filter and which passes the same total power as the filter (given a flat spectrum input such as white noise or an impulse).

Suppose the power density of an arbitrary digital filter is y(f)=|F (f)|^2, the discrete form is Y(f_i)=|F (f_i)|^2, then

where Pm(Y) is the peak value of Y(f_i).

Below is an application of estimating the fb parameter for complex morlet wavelet filter bank to best fit the Glasberg&Moore 1990 auditory filter bank ERB data.

Download: here.

Please give reference if you want to use the code.

My paper stack

2008-01-28T11:16:00.001-06:00

http://www.citeulike.org/user/kelvinchufei

Joint Peristimulus Time Histogram

2008-01-25T15:08:00.000-06:00

Google Scholar Usage Tips

2008-01-25T13:12:00.001-06:00

Google Scholar Citation Export

2008-01-25T13:09:00.000-06:00

1. Go to Google Scholar
2. Click on the Scholar Preferences link to the right of the Scholar search box.
3. Scroll to the bottom of the Scholar Preferences page, and in the 'Bibliography Manager' section, click the radio button next to 'Show link to import citations into' and choose the citation manager format you prefer: BibTeX, EndNote, RefMan, etc.
4. Click the "Save Preferences" button. That's it!

Google Scholar + CiteULike

2008-01-25T13:03:00.000-06:00

CiteULike is a free online service to organise references to academic papers of interest and share them with others. You can create your own library of papers and put tags on them. I feel this is extremely useful as the number of papers I've read increases rapidly.

Link: http://www.citeulike.org/

Tutorial: http://www.citeulike.org/post

You can first use google scholar to find the paper you need then get into the corresponding abstract page, then click the citeulike button, then it's done! You can put as many tags as you want and keep track of all the papers you have read!

Envelop and Temporal Fine Structure of Speech Signal

2008-01-25T12:00:00.000-06:00

We can use hilbert transform to extract the envelop and temporal fine structure of speech signal.

Ref: Chris Shen, Zach Smith, Andrew Oxenham, and Bertrand Delgutte 2006; Gaetan Gilbert and Christian Lorenzi 2006;

Here is my implementation in Matlab R2007b: Download

Please give reference if you want to use the code.

Phonemes in English

2008-01-25T11:25:00.002-06:00

Phoneme Spectral Cues

2008-01-25T11:25:00.001-06:00

How to read a spectrogram?

2008-01-25T11:22:00.000-06:00

Wavelet scalogram using complex morlet wavelet in Matlab R2007b

2008-01-25T11:10:00.001-06:00

In order to see the frequency formants in speech signal on a wavelet scalogram, usually a time window of certain length is needed. Just like in STFT spectogram, depending on the size of the analysis window, different levels of frequency/time resolution are achieved. A long window resolves frequency at the expense of time—the result is a narrow band spectrogram, which reveals individual harmonics (component frequencies), but smears together adjacent 'moments'. If a short analysis window is used, adjacent harmonics are smeared together, but with better time resolution. The result is a wide band spectrogram in which individual pitch periods appear as vertical lines (or striations), with formant structure.

However, the Morlet wavelet function (wname='morl') included in Matlab R2007b wavelet toolbox doesn't have a bandwidth varying parameter.

In Matlab R2007b, the morlet function is defined as:

So we will have to use complex morlet wavelet in which a bandwidth parameter is included:

where fb is a bandwidth parameter, fc is a wavelet center frequency. We can see that the Morlet function above is somehow a form of complex morlet wavelet where fb=2, fc=5/2pi.

For more formal definitions, please see http://en.wikipedia.org/wiki/Morlet_wavelet

In matlab, the continuous wavelet transform using complex morlet wavelet is realized by:

wname = ['cmor',fb,'-',fc];
COEFS = cwt(S,SCALES,'wname');

Here is my script for getting a wavelet scalogram using complex morlet wavelet.Download

Please give reference if you want to use the code.

Middle Ear Filter

2008-01-25T09:47:00.000-06:00

Matlab realization

My Script file: here

Please give reference if you want to use the code.

This old webpage will be reused

2008-01-25T09:17:00.000-06:00

This old webpage would be reused to keep track of my research updates.

Bash and Bash Scripts

2007-05-18T09:06:00.000-06:00

http://www-128.ibm.com/developerworks/library/l-bash2.html

SDVL 64bit Linux Cluster Use

2007-05-18T09:04:00.002-06:00

Netfinity Linux Cluster Use

2007-05-18T09:04:00.001-06:00

Qsub Sample Script

2007-05-18T09:03:00.000-06:00

Qsub Manual Page

2007-05-18T09:02:00.000-06:00

How old are you, rat?

2007-05-16T09:18:00.001-06:00

How to install RTXI with Xenomai on Fedora Core 6

2007-05-14T13:27:00.000-06:00

Installing RTXI requires several software packages to work in concert.

On each of the following pages we will provide detailed instructions on installing this software.

You can get an overview of the software installation from: [http://rtxi.sourceforge.net RTXI at sourceforge]

We used to used Ubuntu 6.06 + RTAI + Comedi + RTXI as a system structure of our dynamic clamp.

But now we've shifted to Fedora Core 6 + Xenomai + RTXI as a more concise and user-friedly combination.

== Preparation ==

You will need:

1. [http://fedoraproject.org/wiki/ Fedora Core 6] as the base linux operating system.

2. [http://www.xenomai.org/ Xenomai] provides the realtime layer.

3. [http://www.rtxi.org RTXI] acts as the main application of the dynamic clamp.

== Install Fedora Core 6==

Usually you can just insert the installation cd and follow the instructions during installation.

A detailed instruction manual can be found at [http://www.informit.com/articles/article.asp?p=169466&rl=1 here].

Sometimes you may need partition before installing fedora if you want to dual-boot systems. You can use many different types of partition softwares in Windows to carry out this job. For example, [http://www.symantec.com/partitionmagic/ Symantec PartitionMagic].

== Install Xenomai ==

1.0 Download the linux base kernel

----

First you need to download Xenomai [http://www.xenomai.org/ here]. Extract the files, for example, into /usr/src/xenomai-2.1.

Then come to the linux kernel. You can find linux base kernel [http://www.kernel.org here]. For example, if you are using 2.6.20 kernel, you may download the kernel and extract it into /usr/src/linux-2.6.20. Then you should make a symbol link from that directory to /usr/src/linux:

'''$ su -'''

'''$ ln -s /usr/src/linux-2.6.20 /usr/src/linux'''

1.1 Preparing the target kernel

----

Xenomai provides a real-time sub-system seamlessly integrated to Linux, therefore the first step is to build it as part of the target kernel. To this end, scripts/prepare-kernel.sh is a shell script which sets up the target kernel properly. The syntax is as follows:

'''$ scripts/prepare-kernel.sh --linux= [--adeos=] [--arch=]'''

--linux specifies the path of the target kernel source tree. Such kernel tree being configured or not makes no difference and is valid either way.

--adeos specifies the path of the Adeos patch to apply against the kernel tree. This parameter can be omitted if Adeos has already been patched in. In any case, the script will not try to apply it again whenever a former patch is detected.

--arch tells the script about the target architecture. If unspecified, the build system architecture is detected and suggested as a reasonable default.

Introduction about ADEOS can be found [http://home.gna.org/adeos/ here]. You can download the corresponding adeos patch with your linux kernel version.

For instance, the following command would prepare the Linux tree located at /usr/src/linux-2.6.20 in order to include the Xenomai support:

'''$ su -'''

'''$ cd xenomai-2.1'''

'''$ scripts/prepare-kernel.sh --linux=/usr/src/linux --adeos=/tmp/adeos-ipipe-2.6.20-i386-1.0-09.patch --arch=i386'''

Note: The script will infer the location of the Xenomai kernel code from its own location within the Xenomai source tree. In other words, if /usr/src/xenomai-2.1/script/prepare-kernel.sh is executing, then Xenomai's kernel support available from /usr/src/xenomai-2.1/ksrc will be bound to the target kernel.

1.2 Configuring and building the target kernel

----------------------------------------------

Once the target kernel has been prepared, all Xenomai configuration options are available from the "Real-time subsystem" toplevel menu.

There are several configure options that cause large latencies; they should be avoided. The TROUBLESHOOTING file identifies them and explains the issues with their use. Once configured, the kernel should be built as usual.

In the following steps, suppose $xenomai_root = path to the Xenomai sources and $build_root = path to a clean build directory.

'''$ su -'''

'''$ cd /usr/src/linux'''

'''$ make xconfig/gconfig/menuconfig''' # remeber to select the kernel and Xenomai options

'''$ make bzImage modules''' # then install as needed

'''$ mkdir $build_root && cd $build_root'''

'''$ $xenomai_root/configure --enable-x86-sep'''

'''$ make install'''

You can now restart the machine and boot into the new kernel.

* [[undefined reference to 'smp_processor_id_hw' error]]

== Install RTXI ==

Download the RTXI from CVS server [http://sourceforge.net/cvs/?group_id=118411 here].

'''$ cvs -z3 -d:pserver:anonymous@rtxi.cvs.sourceforge.net:/cvsroot/rtxi co -P rtxi'''

Extract it to /usr/src/rtxi.

'''$ su -'''

'''$ cd /usr/src/rtxi'''

'''$ ./autogen.sh'''

'''$ ./configure --enable-xenomai --disable-comedi --prefix=/usr/src/rtxi --enable-ni'''

'''$ make install'''

You should be able to run:

'''$ rtxi'''

How to install Graphical User Interface for yum

2007-03-21T16:31:00.001-06:00

Yumex is a good thing.

$yum -y install yumex

How to config remote desktop on Fedora Core 6

2007-03-21T15:24:00.000-06:00

How to configure:

"Desktop" -> "Preferences" -> "Remote Desktop"
in "Remote Desktop Preferences"

"Sharing" ->
"Allow other users to view your desktop" (Checked)
"Allow other users to control your desktop" (Checked)

"Security" ->
"Ask you for confirmation" (Un-Checked)
"Require the user to enter this password:" (Checked)
Password: Specify the password

How to connect into remote Fedora desktop:

Use VNCViwer to connect.

From Fedora, execute
vncviewer -fullscreen 192.168.0.1:0

From Windows,
http://fresh.t-systems-sfr.com/cgi-bin/warex?linux/src/vnc-4_1_1-x86_win32_viewer.zip:a/vnc-4_1_1-x86_win32_viewer.exe

How to install VNC server:
"Desktop"->"Applications"->"Add/Remove Package"

How to use CVS on Fedora Core 6

2007-03-20T14:35:00.000-06:00

First thing is to log into the CVS server

$export CVSROOT=:pserver:cvs_usrname@the_server_name:/home/cvsroot
$cvs login

1. How to Import your project to CVS server
Enter the project directory you've got, such as "cvstest"
$cd cvstest
$cvs import -m "this is a cvstest project" cvstest start v_0_0_1
after -m is something you want to note on this version. "cvstest" is the project name. "start" is the vendor_tag, "v_0_0_1" is release_tag.
(cvs import -m "write some comments here" project_name vendor_tag release_tag)

2. How to Download your project to a new machine
$cvs checkout cvstest

3. How to Update your project to newest version (you've got a older copy)
$cd cvstest
$cvs update
or $cvs update filename

4. How to See what's new in CVS server
$cd cvstest
$cvs status

5. How to Add a new file to the project
$cvs add new_file
Note: for pictures or other non-text items, you need to use "cvs add -kb" to add file with binary mode, such as:
$cvs add -kb new_file.gif
$cvs add -kb readme.doc

6. How to Remove a file from the project
$cvs rm file_name

7. How to Finally Submit your edit
After you edit, add or delete files, don't forget to commit the action.
$cvs commit -m "write some comments here" file_name

8. Abbreviations:
commit=>ci; update=>up; checkout=>co/get; remove=>rm;
For example, "$cvs commit" can be writtern as "$cvs ci"

9. How to Recover a previous version
(1) $cvs log
Find the version you want to recover, for example, version 1.2
(2) $cvs update -p -r1.2 file_name >file_name
or $cvs update -p -r1.2
Then the old version 1.2 is coming back.

How to setup CVS server on Fedora Core 6

2007-03-20T13:24:00.000-06:00

Login as "root":

(1) Get xinetd package installed

(2) Get cvs package installed

(3) Create a group CVS

(4) Create a user cvsroot and assign it to CVS group, the home directory for user cvsroot should now be at /home/cvsroot.

(5) $gedit /etc/services
and look for "cvspserver". There should be 2 lines like:
cvspserver 2401/tcp # CVS client/server operations
cvspserver 2401/udp # CVS client/server operations
If there are no such lines, add them into /etc/services.

(6) In /etc/xinet.d/, look for file "cvs", if there is no such file, create it.
Edit the file /etc/xinet.d/cvs and make it look like:
# default: off
# description: The CVS service can record the history of your source \
# files. CVS stores all the versions of a file in a single \
# file in a clever way that only stores the differences \
# between versions.
service cvspserver
{
port = 2401
socket_type = stream
protocol = tcp
wait = no
user = root
server = /usr/bin/cvs
server_args = -f --allow-root=/home/cvsroot pserver
}

(7) $/etc/init.d/xinetd restart

(8) In "System"->"Administration"->"Security Level and Firewall", add firewall exception of port "2401" corresponding to "cvspserver" service, and also remember to "Modify SELinux policy"->"SELinux Service Protection"->"Disable ELinux Service Protection for cvs".
This step is proved to be very important!

Login as "cvsroot":

(9) $cvs -d /home/cvsroot init

(10) $cd /home/cvsroot

(11) $chmod 771 CVSROOT
CVSROOT is a directory automatically generated under /home/cvsroot.

At this time, all local real system users can log into CVS server with their own passwords.
If you want to disable local real system users' login, you can do by:

(12) $ chmod 644 /home/cvsroot/CVSROOT/config
$ vi $CVSROOT/CVSROOT/config
Here you need to uncomment the line where #SystemAuth = yes. And set SystemAuth = no.

Then now your system users cannot use CVS service with their own system passwords, except you assign them manually in file /home/cvsroot/CVSROOT/passwd.

(13) To control user access in CVS, there are three files in /home/cvsroot/CVSROOT/, respectively named passwd, readers, writers. The file readers and writers are not there by default, so you have to create them manually.

In file "passwd", ithere stores the shadow files for username and password, following the rule:
{cvs usrname}:[encrypted pwd]:[equivalent system usrname]
where the second field is encrypted with crypt(3), just like in /etc/shadow.
The thrid field equivalent system usrname means the corresponding system priviledge the cvs usrname has.

In file "readers", there stores the list of cvs usrnames of whom you want to enable to only have read acess.

Similarly, in file "writers", there stores the list of cvs usrnames of whom you want to enable to only have read acess.

Now, let us suppose we have CVS users "henry", "betty", "anonymous". And we don't want to give them system access, which means the corresponding system priviledge the cvs usrnames have are all null. So, we need to create an equivalent system user, which names "cvspub", who has no system priviledge at all.

$ useradd -g CVS -M -s /sbin/nologin cvspub

Then edit file "passwd", make it like:

henry:$fajkdpaieje:cvspub
betty:fjkal;ffjieinfn/:cvspub
anonymous::cvspub

Now suppose we want to let henry has write acess while betty and anonymous only have read acess.
Then edit file "readers", make it like:
betty
anonymous

And edit file "writers", make it like:
henry

Then it is all finished.

(14) Test login:
$export CVSROOT=:pserver:henry@the_server_name:/home/cvsroot
$ cvs login
(System will ask for password for henry)
Passwd:xxxxxxxx
There should be no error when you have successfully login to the CVS server.

Problem:
cvs [login aborted]: unrecognized auth response from brachial.umn.edu: cvs pserver: cannot open /home/cvsroot/CVSROOT/config: Permission denied

Solution:
In "System"->"Administration"->"Security Level and Firewall", add firewall exception of port "2401" corresponding to "cvspserver" service, and also remember to "Modify SELinux policy"->"SELinux Service Protection"->"Disable ELinux Service Protection for cvs".
This step is proved to be very important!

Problem:
cvs import: cannot open /root/.cvsignore: Permission denied

Solution:
Add "-f" to the arguments for cvs in /etc/xinet.d/cvs:
server_args = -f --allow-root=/home/cvsroot pserver

Linux Group and User Account Administration

2007-03-20T12:09:00.001-06:00

How to setup CVS (English)

2007-03-20T11:52:00.000-06:00

http://www.cs.umb.edu/~srevilak/cvs.html
http://owen.sj.ca.us/rkowen/howto/cvs.html

CVS manual (Chinese)

2007-03-20T11:50:00.001-06:00

How to setup CVS (Chinese)

2007-03-20T11:46:00.000-06:00

http://www.pgsqldb.org/cvstutorials.html
http://www.linuxmine.com/type1070.html

Integrating PHP with Apache on Windows

2007-03-14T10:45:00.000-06:00

1. Open up the Apache server configure file

2. Find a section "LoadModule"
At the end of this section, add the following:

LoadModule php5_module C:/php/php5apache2_2.dll (or php5apache2.dll)

3. Next look for the section "AddType"
At the end, add the following lines:

AddType application/x-httpd-php .php .html
PHPIniDir "C:/php"

4. Restart Apache Server.

Use what command to get the library directory

2007-03-13T15:33:00.000-06:00

rpm -q -l qt-devel

Posix enabled RTXI install

2007-03-13T14:15:00.000-06:00

1.Install the posix patch you need to get rtxi from cvs

cvs -d:pserver:anonymous@rtxi.cvs.sourceforge.net:/cvsroot/rtxi login

cvs -z3 -d:pserver:anonymous@rtxi.cvs.sourceforge.net:/cvsroot/rtxi co -P RTXI

2.Enter that directory and apply the patch:

patch -p1 < path/of/the/patch(posix.diff)

3.Compile rtxi, and make sure you give it the --enable-posix flag when you configure.

./autogen.sh

./configure --enable-posix

undefined reference to `smp_processor_id_hw' error

2007-03-12T12:11:00.001-06:00

Problem:
When we are compiliing the new kernel on Fedora Core 6, it returns the error of

LD .tmp_vmlinux1
arch/i386/mach-generic/built-in.o: In function `send_IPI_allbutself':
/usr/src/linux/include/asm/mach-summit/mach_ipi.h:14: undefined reference to `smp_processor_id_hw'
arch/i386/mach-generic/built- in.o: In function `send_IPI_allbutself':
/usr/src/linux/include/asm/mach-bigsmp/mach_ipi.h:14: undefined reference to `smp_processor_id_hw'
arch/i386/mach-generic/built-in.o: In function `target_cpus':
/usr/src/linux/include/asm/mach-es7000/mach_apic.h:19: undefined reference to `smp_processor_id_hw'
/usr/src/linux/include/asm/mach-es7000/mach_apic.h:19: undefined reference to `smp_processor_id_hw'
arch/i386/mach-generic/built- in.o: In function `send_IPI_allbutself':
/usr/src/linux/include/asm/mach-es7000/mach_ipi.h:14: undefined reference to `smp_processor_id_hw'
arch/i386/mach-generic/built-in.o:/usr/src/linux/include/asm/mach-default/mach_ipi.h:22: more undefined references to `smp_processor_id_hw' follow
make: *** [.tmp_vmlinux1] Error 1

Solution:
You need to change the sub-architecture type to "PC-compatible" you currently have it on generic.

After Fedora pakcage update, cursor disappears

2007-03-09T15:37:00.001-06:00

Solution:

As root user, edit /etc/X11/xorg.conf and add:

Option "HWCursor" "off"

to the Device section

save the file and restart X

How to compile a kernel on Redhat/Fedora

2007-03-09T12:53:00.000-06:00

1. Download the correpsonding kernel you want to use from www.kernel.org
2. Unzip the source files to /usr/src/linux
3. Make sure you have all the necessary lib file and devtools, for example, gcc, glibc, qt3/4, etc...
4. Execute: $ cp /boot/config-'uname -r' ./.config
5. Make xconfig
6. Make all
7. Make modules_install
8. Make install

You may want to change Grub bootfile also: /boot/grub/menu.lst

Edit /etc/X11/xorg.conf

2007-03-06T16:11:00.000-06:00

# Xorg configuration created by pyxf86config

Section "ServerLayout"
Identifier "Default Layout"
Screen 0 "Screen0" 0 0
InputDevice "Keyboard0" "CoreKeyboard"
EndSection

Section "InputDevice"
Identifier "Keyboard0"
Driver "kbd"
Option "XkbModel" "pc105"
Option "XkbLayout" "us"
EndSection

Section "Device"
Identifier "NVIDIA 6100"
Driver "vesa"
Option "HWCursor" "off"
EndSection

Section "Monitor"
Identifier "Generic Monitor"
VendorName "Dell"
ModelName "UltraSync 17"
HorizSync 31.5-48.5
VertRefresh 50-70
EndSection

Section "Screen"
Identifier "Screen0"
Device "NVIDIA 6100"
Monitor "Generic Monitor"
DefaultDepth 24
SubSection "Display"
Viewport 0 0
Depth 24
Modes "1280x1024"
EndSubSection
EndSection

RPM HOWTO

2007-03-06T15:46:00.000-06:00

Quick Guide to Red Hat's Package Manager (RPM)

2007-03-06T15:43:00.000-06:00

How to backup Ubuntu

2007-03-06T13:20:00.000-06:00

How to move RTXI from old machine to a new one

2007-03-01T13:40:00.000-06:00

1. RTAI-patched Linux
Copy the linux source file from old machine(which should already be RTAI patched)
More See http://kelvinchufei.blogspot.com/2006/12/build-real-time-kernel-on-ubuntu-606.html

Note:You should also directly copy and install the kernel_img.deb and kernel_header.deb from your old machine.

2. RTAI
Copy the RTAI source file from old machine or you can download RTAI_vulcano from RTAI official site.

(Note: Version 3.5 hasn't been test yet but version 3.4 is proved to be working)

3. RTXI
Copy the RTXI source file from old machine.
Download automake1.9 debian package from Internet and install it.
Execute:
./autogen.sh
./configure --enable-debug --prefix=/usr/src/rtxi
make
make install

4. "Theoratically" you don't need to install comedi/comedilib. But we need further investigation.

What is x86-64

2007-03-01T13:36:00.000-06:00

Add scripts to run when booting up Linux

2007-03-01T13:29:00.000-06:00

If you want your command to run at boot time, just place it in /etc/rc.local without the sudo in front of it, above "exit 0". You can test it by running the command:

sudo /etc/init.d/rc.local start.

Linux Startup Sequence & Scripts

2007-03-01T13:28:00.001-06:00

Potential X problem with Ubuntu 6.06

2007-03-01T13:15:00.000-06:00

Problem:

X Error: BadDevice, invalid or uninitialized input device 166
Major opcode: 146
Minor opcode: 3
Resource id: 0x0
Failed to open device
X Error: BadDevice, invalid or uninitialized input device 166
Major opcode: 146
Minor opcode: 3
Resource id: 0x0
Failed to open device

Solution:

Go into xorg.conf and comment out the following lines:

Code:
# InputDevice "stylus" "SendCoreEvents"
# InputDevice "cursor" "SendCoreEvents"
# InputDevice "eraser" "SendCoreEvents"And:

Code:
#Section "InputDevice"
#
# # /dev/input/event
# # for USB
# Identifier "stylus"
# Driver "wacom"
# Option "Device" "/dev/wacom" # Change to
# Option "Type" "stylus"
# Option "ForceDevice" "ISDV4" # Tablet PC ONLY
#EndSection
#
#Section "InputDevice"
#
# # /dev/input/event
# # for USB
# Identifier "eraser"
# Driver "wacom"
# Option "Device" "/dev/wacom" # Change to
# Option "Type" "eraser"
# Option "ForceDevice" "ISDV4" # Tablet PC ONLY
#EndSection
#
#Section "InputDevice"
#
# # /dev/input/event
# # for USB
# Identifier "cursor"
# Driver "wacom"
# Option "Device" "/dev/wacom" # Change to
# Option "Type" "cursor"
# Option "ForceDevice" "ISDV4" # Tablet PC ONLY
#EndSection

How to install RTXI on Ubuntu 6.06 (1)

2007-03-01T12:59:00.000-06:00

1. Make sure the RTAI kernel is working (i.e. you can load realtime modules and do the latency test successfully.)
See http://kelvinchufei.blogspot.com/2006/12/build-real-time-kernel-on-ubuntu-606.html
See also http://kelvinchufei.blogspot.com/2006/12/myscripts-memo.html

2. Get the latest version of "automake" such as version 1.9
(don't use the one Ubuntu provided by default thru apt-get, but instead find a seperate .deb package)

3. ./autogen.sh

4. ./configure --enable-debug --prefix=/usr/src/rtxi

5. make

6. make install

How to install .deb packages

2007-03-01T12:51:00.000-06:00

1. First login as root
2. dpkg -i yourpackage.deb
3. Done

How to create Image (ISO) files from folders in Ubuntu 6.06

2007-02-27T14:26:00.000-06:00

$ mkisofs -r -o file.iso /location_of_folder/

How to check MD5 checksum of files in Ubuntu 6.06

2007-02-27T14:24:00.000-06:00

Assumed that file.iso and file.iso.md5 are in the same folder
then execute:
$ md5sum -c file.iso.md5

How to install Matlab R2006b (Unix) on Ubuntu 6.06 / Fedora Core 6

2007-02-27T14:04:00.000-06:00

1.Download the iso files for Matlab R2006b for Unix
2.Mount the first iso by:
(a)sudo mkdir /media/iso
(b)sudo modprobe loop
(c)sudo mount file1.iso /media/iso/ -t iso9660 -o loop
(if you use CDs, you can directly find /media/Mathworks_2006B1/)
3.Make your $Matlab Directory by:
(a)mkdir /usr/local/matlab73
(b)cd /usr/local/matlab73
4.Copy your license file to this directory
(a)cp /home/kelvin/license.dat ./
5.Start the installation by:
(a)sudo sh /media/iso/install
(b)Do the installation configurations you want
(c)Continue following the instructions
6. When the system reminds you of changing to disc 2,
(a)sudo umount /media/iso
(b)sudo mount file2.iso /media/iso/ -t iso9660 -o loop
7. Continue installation and do the same with disc 3.
8. Finish installation.

FAQs:
1.Problem:
Exception in thread "AWT-EventQueue-0" java.lang.InternalError: Current locale is not supported
at sun.awt.motif.MWindowPeer.pSetTitle(Native Method)
at sun.awt.motif.MWindowPeer.init(Unknown Source)
at sun.awt.motif.MFramePeer.(Unknown Source)
at sun.awt.motif.MToolkit.createFrame(Unknown Source)
...........

Solution:
$locale
and if its not en_US.utf8 then change it like this:

export LANG=en_US.UTF-8

Add this line to .bashrc too....and enjoy Matlab..

2.Problem: When trying to run matlab, it does not want to start. Error message:
/usr/local/matlab73/bin/glnx86/MATLAB: error while loading shared libraries: /usr/local/matlab73/bin/glnx86/libmwlapack.so: cannot restore segment prot after reloc: Permission denied

Solution:
chcon -t texrel_shlib_t /usr/local/matlab73/bin/glnx86/*.so
or disable "SELinux" from system security panel.

3.Problem:
/tmp/12680tmwinstall/update/bin/glnx86/xsetup: error while loading shared libraries: libXp.so.6: cannot open shared object file: No such file or directory

Solution:
Remember to install xorg-x11-libXp package.

4.Problem: Matlab requires a Terminal to start.If I create a launcher with the command Matlab in the Gnome-menu I just see the splash-screen and then nothing.

Solution:
Create a luncher with "matlab -desktop"

A control module that can pause/unpause other function modules

2007-02-27T12:36:00.000-06:00

me: hi
now I'm looking at the RT::Thread::setActive()
do you have any example for how to doing with this function?

jbetten: for a model to pause itself you just call setActive(false)
to pause another model, you just do modelPtr->setActive(false)

me: how can I find the modelPtr
I can understand that each module must have a pointer or handler

jbetten: if you know the model ID you can use Settings::Manager::getInstance()->getObject()
you then need to up cast it

me: what do you mean by model ID

jbetten: every object has a unique ID

me: where is it defined

jbetten: it is in the caption of each default gui model

me: within the source code I couldn't expect what ID is allocated

jbetten: no

me: what I can do is providing a GUI
and let the user to input the model IDs they want to pause
is that what you mean?

jbetten: yeah

RTXI new custom modules

2007-02-26T17:36:00.000-06:00

For the past 3 weeks, I was writing modules for RTXI. The finished modules include:

(1)SpikeRateController
Input: State
Output: Current

(2)PSG (synapitic model)
Input: State & Vm
Output: Current

(3)PlayBack
Input: File (Structure: TIME Vm)
Output: Vm

(4)EventPlayBack
Input: File (Structure: TIME Cell#0-3)
Output: State

(5)DualCellSyn (Discarded)

Read the parameter from GUI in RTXI

2007-02-26T17:29:00.000-06:00

On 2/16/07, Fei Chu < kelvinchufei@gmail.com> wrote:
Hi, Jonathan
Now we are writing the playback module which needs to read a data file. But when we need to specify the filename, we don't know how to read the string from the Default GUI. It seems that there are only datatypes for double, uinteger and integer. What do you suggest us to do?
Thanks.
Kelvin

On 2/17/07, Jonathan Bettencourt < jbetten@gmail.com> wrote:
Hey,
If you don't specify a type none is enforced, so you effectively have a string type.
-JB

On 2/22/07, Fei Chu wrote:
Hi, Jonathan
I defined
char *FileName;
But when
FileName = getParameter("FileName"));
It reports error
invalid conversion from 'const char*' to 'char*'
What should I do with this converting?

Thanks,
Kelvin

On 2/22/07, Jonathan Bettencourt < jbetten@gmail.com> wrote:

I think you are getting that problem because getParameter returns a QString instead of a char * what you probably want is something like:

char FileName[256];
memset(FileName,0,sizeof(FileName));
strncpy(FileName,getParameter("FileName").latin1(),255);

or just use getParameter directly whenever you need the filename:

int fd = open(getParameter("FileName").latin1(),"rw");

-JB

Default GUI patch of RTXI @ Feb.26th

2007-02-26T17:25:00.000-06:00

On 2/24/07, Jonathan Bettencourt wrote:
> Hey,
>
> I'm going to do some work on DefaultGUIModel tomorrow, namely:
> 1) Have Applied Parameters appear as red, like RTLDC.
> 2) Remove the "Refresh" button and have the GUI auto refresh every second or
> so.
> 3) Add a pause button.
> 4) Make DefaultGUIModels models load paused.
> 5) Make DefaultGUIModel::update() run in non-realtime, while the model is
> paused.
> 6) Stop having DefaultGUIModels automatically connected to the DAQ when they
> load.
>
> Also if you remind me I can send you a patch to fix the connector plugin so
> it behaves more correctly. And I'll try to get you a patch that makes
> RT::Thread::setActive() public, so it can be called by other models.
>
> With RT::Thread::setActive() made public you should be able to unpause a
> bunch of models "simultaneously", and by that I mean in realtime with an
> RT::Event.
>
> -JB

From: Jonathan Bettencourt
Date: Feb 24, 2007 10:02 AM
Subject: Re: Synchronization
To: Theoden Netoff

Hey,

Here is a patch that fixes the connector plugin and makes the changes
I mentioned in the previous email. I'm not sure how cleanly it will
apply but it is worth a shot.

To Apply:

1) Backup your existing RTXI directory.
cd /usr/src
cp -R rtxi rtxi.old

2) Apply the patch.
patch -p0 < /the/path/to/the/patch.connector_and_defaultgui

3) See if the patch applied cleanly, i.e. the output of the previous
step didn't mention and failures. If it fails email me the output of patch and revert to using your old rtxi tree. Otherwise run make install and you should be good to go.

-JB

How to generate correlated gaussian noises

2007-02-26T17:05:00.000-06:00

Theory http://www.columbia.edu/~mh2078/MCS04/MCS_framework_FEegs.pdf

So here is the Matlab code for generating 3 channels of correlated gaussian noise:

cwn.m

How to Write RTXI Modules

2007-02-23T12:02:00.000-06:00

The source file of a typical RTXI module may include three parts: Makefile.am, YourMod.cpp, YourMod.h

Take mimic module as an example.

1. The Makefile.am is like

CLEANFILES = *~
DISTCLEANFILES =
MAINTAINERCLEANFILES = Makefile.in

include $(top_srcdir)/.buildvars

pkglib_LTLIBRARIES = mimic.la

mimic_la_LDFLAGS = -module -avoid-version

mimic_la_SOURCES = \
mimic.h \
mimic.cpp

Usually what you need to do is just change all "mimic" to YourMod.

Of course, don't forget to use ./autogen.sh and ./configure --enable-debug --prefix=/usr/src/rtxi to generate the real Makefile before you can compile.
See http://kelvinchufei.blogspot.com/2007/02/how-to-make-new-module-for-rtxi.html.

2. The Header file mimic.h is like this:

#include

//Here uses the default gui, of course if you want you can write your own gui and include it here

class Mimic : public DefaultGUIModel
{

public:

Mimic(void);
virtual ~Mimic(void);

virtual void execute(void);

protected:

virtual void update(DefaultGUIModel::update_flags_t);

private:

//Here of course you can define your own functions
//void SetPSG(double *, double, double, double);

double gain;
double offset;

};

3. The main mimic.cpp is like this:

#include

extern "C" Plugin::Object *createRTXIPlugin(void) {
return new Mimic();
}

static DefaultGUIModel::variable_t vars[] = {
{
"Vin",//Label name in GUI
"",//Here is comments you want to put
DefaultGUIModel::INPUT,//Type of the GUI element: INPUT,OUTPUT,PARAMETER,STATE
},
{
"Vout",
"",
DefaultGUIModel::OUTPUT,
},
{
"Gain",
"Scaling of the copied signal",
DefaultGUIModel::PARAMETER | DefaultGUIModel::DOUBLE,
//DateType: DOUBLE, INTEGER, UINTEGER
//When defined as DefaultGUIModel::PARAMETER directly, the input type is treated as string
},
{
"Offset",
"Offset of the copied signal",
DefaultGUIModel::PARAMETER | DefaultGUIModel::DOUBLE,
},
};

static size_t num_vars = sizeof(vars)/sizeof(DefaultGUIModel::variable_t);

Mimic::Mimic(void)
: DefaultGUIModel("Mimic",::vars,::num_vars), gain(1.0), offset(0.0)
//Initialize the parameters
{
update(INIT);
refresh();
}

Mimic::~Mimic(void) {}

void Mimic::execute(void) {
output(0) = gain*input(0)+offset;

//Main function, where you can use input(0) as first input etc and use output(0) as the first output.
}

void Mimic::update(DefaultGUIModel::update_flags_t flag) {
switch(flag) {
case INIT:
setParameter("Gain",gain);
setParameter("Offset",offset);
//remember to relate all the parameters with the GUI elements
break;
case MODIFY:
gain = getParameter("Gain").toDouble();
offset = getParameter("Offset").toDouble();
//remember to relate all the parameters with the GUI elements
break;
default:
break;
}
}

How to compile a new module for RTXI

2007-02-15T13:22:00.000-06:00

1. You need to have a Makefile.am YourMod.cpp YourMod.h in your /rtxi/models/YourMod directory
(The structure of a Makefile can refer to Makefiles in other module directories. Or you can simply replace the name of other modules in that Makefile into your module name and use it.)

2.Change /rtxi/configure.in and after AC_CONFIG_FILES(), add /models/YourMod/Makefile into it.

3.Change /rtxi/models/Makefile.am and after SUBDIR=, add YourMod.

4.Run ./autogen.sh (create makefile.in)

5.Run ./configure --enable-debug --prefix=/usr/src/rtxi(Your rtxi directory)

6. Make & Make install

7. Find your module plugin in /rtxi/lib/rtxi/

Agar Recipe

2007-01-30T16:31:00.000-06:00

Sigma Agar #A-9915

1. 125 mL of water
2. heat till boiling
3. add 5g of agar powder and stir
4. when it becomes transparent and smooth, pour into petri dishes cover asap.

The amount makes about 3 petri dishes (100*15mm)

In my experience this is a slightly viscous mixture, I used a bit less
myself last time. Also, this makes so much agar, it dries out before you can really use it all. The last batch lasted us almost 9 months.

Agar Recipe

2007-01-25T16:31:00.000-06:00

Sigma Agar #A-9915

500 mL of water, add 20g of agar powder
Stir and heat(slowly) till boiling, pour into petri dishes cover asap.

makes about 8 petri dishes

In my experience this is a slightly viscous mixture, I used a bit less myself last time. Also, this makes so much agar, it dries out before you can really use it all. The last batch lasted us almost 9 months.

Trainings required by univeristy to perform animal experiments

2007-01-25T14:16:00.000-06:00

1. Radiation Training(Lab security www.ahcrbms.umn.edu)
2. IACUC Animal and chemical training (Protocol http://www.research.umn.edu/iacuc/training/)
3. Animal Raising/Micro-isolator Traing (RAR http://www.ahc.umn.edu/rar/)

Osmolarity Test

2007-01-25T13:35:00.000-06:00

ACSF ~ 300mOSM/kg
ICF ~ 270-290mOSM/kg

1. Put a sample disc on the tray
2. Put 10ul sample solution on there
3. Close the test chamber
4. Wait and read
5. Wipe it clean

ICF Recipe

2007-01-25T13:23:00.000-06:00

Name Final-Conc Stock-Conc Quantity for 50ml ICF
K-gluconate 120mM powder 1.4052g
KCL 10mM 1M/L(3.728g/50ml) 0.5ml
Hepes 10mM powder 0.1192g

Then quickly add

Mg-ATP 4mM powder 0.1188g
Tris-GTP 0.3mM 10mM/L 1.5ml
Na2-Phoscreatine 10mM powder 0.1690g
Creatine Kinase 20u/ml 1000u/ml 1ml

Use the pH meter to bring it to PH 7.25 (Use Potassium Hydroxide 1N solution)

Vacuum suck to filter the solution. (2um) because our pipette tip is 1-2um in diameter.

Check Osmolarity which should be 270~290mOSM/kg.(Use sucrose to increase, water to decrease). (ACSF osmolarity ~300mOSM/kg)

Micro-pipette Pulling Cycle / Resistence Test

2007-01-25T12:19:00.000-06:00

Micro-pipette Pulling Cycle / Resistence Test is used to get a good parameter combination for future reproducible use.

Usually it is conducted by the following procedure:
1. Set up certain combination of pulling parameters to ensure the cycle times are 3 times and the length of time is about 10s
2. Inject 1* ACSF / ICF into micro-syringe and make sure there is no bubble in the syringe
(a) take 50ml 5* ACSF
(b) bring to 250ml solution
(c) Add 500ul CaCl2 of 1mol/L concentration
(d) Shake it to make it transparent solution without precipitation
3. Connect the tip needle with the syringe and inject a little amount off to expel the air bubbles in the needle
4. Chloride the metal tip of the headstage
5. Inject ACSF / ICF into the pipette to half of the pipette volume (because when the )
6. In the MultiClamp software, first in "Voltage Clamp" mode to balance the offset to zero then go into "Current Clamp" mode to test the "Bridged Resistence".
7. The appropriate resistence range should be from 3-4M ohm.

Working with SPF Animals - RAR facility

2007-01-25T12:08:00.000-06:00

Micro-isolator operation training
http://www.ahc.umn.edu/rar/

1. Turn on Hood (Light/Power)
2. Take on Lab coat / Glove / Sleeve
3. 1 gallon water with 15ml bleacher makes L-cide
4. Spray ("Sodium" labeled, not "DMQ" labeled) the tray (stage, two side glasses, DON'T back filter)
5. Spray the crate
6. Open Lid
7. Decontaminate the Glove again
8. Operation

Contact: Jessica Jones,SPF Rodents ....625-2643

Animal facility address: Jackson Hall 1-107

Different Types of Laboratory Flasks

2007-01-13T15:29:00.001-06:00

Lab Equipment Glossaries

2007-01-13T14:52:00.000-06:00

How to weigh chemicals

2007-01-13T14:05:00.000-06:00

1. Put a weighing paper or bowl on electronic balance;
2. Clear the tare;
3. Put the chemicals on weighing paper or bowl with chemical spatula(if solid) or pipette(if liquid)
4. Weigh it

* You cannot put the extra chemicals(if more than you want) back into the bottle or bucket container.

ACSF 5* Recipe (Artifical Cerebral Spinal Fluid)

2007-01-13T13:53:00.000-06:00

Concentration(5*) MW USE
629mM NaCl 58.442g/M 36.525g(/L)
125mM NaHCO3 84.007g/M 10.501g(/L)
125mM Glucose 180.16g/M 22.525g(/L)
6.25mM NaH2PO4-H2O 137.99g/M 0.8625g(/L) (Sodium phosphate, monobasic H4NaO5P)
12.5mM KCl 74.551g/M 0.932g(/L)
5mM MgCl - 5ml of 1M/L ready solution

* First add 800ml de-ionized water to mix then bring to 1000ml (1 Liter).
** Lable as (1L ACSF 5*, Date & Initial) and store in refrigerator for future use.

Before use:

I. Take 200ml ACSF 5*

II. Bring to 800ml with de-ionzied water

III. Add 200ul of 1M/L CaCl2

(1M/L CaCl2: 11.1g first mixed with 80ml de-ionized water then bring to 100ml because MW=110.98)
Concentration MW USE
0.2mM CaCl2 - 200ul of 1M/L ready solution

IV. Bring to 1000ml (1 Liter)

Final: 1L ACSF

Ref 1: http://www.biocenter.helsinki.fi/bi/physiol/Internal_temp/internal/int_methods/met_ACSF.htm

Ref 2:
http://physics.ucsd.edu/neurophysics/lab/ACSF.pdf

Dynamic Clamp Introduction

2007-01-11T16:19:00.000-06:00

The Dynamic Clamp protocol, developed by Sharp et. al. (J. Neurophysiol. 69, 992-995, 1993) allows "insertion" of simulated membrane conductances in, and/or simulated synapses between biological neurons.The dynamic clamp uses computer simulation to introduce artificial membrane or synaptic conductances into biological neurons and to create hybrid circuits of real and model neurons. In the ten years since it was first developed, the dynamic clamp has become a widely used tool for the study of neural systems at the cellular and circuit levels. This review describes recent state-of-the-art implementations of the dynamic clamp and summarizes insights gained through its use, ranging from the role of voltage-dependent conductances in shaping neuronal activity to the effects of synaptic dynamics on network behavior and the impact of in vivo-like input on neuronal information processing.

The dynamic clamp is a novel method that uses computer simulation to introduce conductances into biological neurons. This method can be used to study the role of various conductances in shaping the activity of single neurons, or neurons within networks. The dynamic clamp can also be used to form circuits from previously unconnected neurons. This approach makes computer simulation an interactive experimental tool, and will be useful in many applications where the role of synaptic strengths and intrinsic properties in neuronal and network dynamics is of interest.

Most influential researchers on the topic of Dynamic Clamp on Scholarpedia.com
http://www.scholarpedia.org/article/Dynamic_Clamp

Hey, Tay! You are on there!

Using the latest RTAI from CVS

2007-01-10T16:58:00.000-06:00

cvs -d:pserver:anonymous@cvs.gna.org:/cvs/rtai co

You can access to see the directories from
http://cvs.gna.org/cvsweb/?cvsroot=rtai

RAMP Value Test

2007-01-10T14:03:00.000-06:00

Before starting to use a new type of glass to pull micro-pipettes, make sure to conduct the RAMP Value Test,which would give an empirical relation to calculate the proper Heat, Pull force, Velocity, Delay, and Pressure.

Operation Procedure:
1. enter a number (such as "3")
2. press "Enter"
3. press "Clear"
4. choose "No"
5. press "1" - Ramp test
6. press "Pull"
7. press "Reset"

Typical Combinations: (may only applicable to Sutter Instruments P-97)

Type:Patch Glass:1.5mm*0.86mm(Sutter Glass BF 150-86-10)
RAMP Test Value: 565
Heat=RAMP Value
Pull=0
Velocity=25
Delay=1
Pressure=500
Loop 4 to 5 times. Pipette will have 2-8 M Ohm Resistance.

Type:Patch Glass:1.5mm*1.10mm(Sutter Glass BF150-110-10)
RAMP Test Value:
Heat=RAMP Value+5
Pull=0
Velocity=45
Time=250
Pressure=500
Loop 3 times. Pipette will have 1-4 M Ohm Resistance.

Type:Micro-injection Glass:1.0mm*0.50mm(Sutter Glass BF100-50-10)
RAMP Test Value: 535
Heat=RAMP Value+15
Pull=60
Velocity=60
Time=250
Pressure=500
Pipette will have 40-80 M Ohm Resistance.

Type:Micro-injection Glass:1.0mm*0.78mm(Sutter Glass BF100-78-10)
RAMP Test Value:
Heat=RAMP Value+5
Pull=90
Velocity=70
Delay=70
Pressure=200
Pipette will have 20-60 M Ohm Resistance.

How to Glue the Chamber

2007-01-10T13:59:00.000-06:00

1. Use the silicon seal in the chamber
2. Put a piece of glass on the chamber then take it off, in order to let the amount of gel on the chamber to be appropriate
3. Put another piece of glass on the chamber
4. Screw to fix the window in the center of the chamber

Instrument Models List

2007-01-09T17:30:00.000-06:00

Micro-pipette Puller : Sutter Instrument P-97
CCD Camera : XC-ST50+ Olympus U-TV0.5XC-3
Microscope: Olympus BX51WI Fixed Stage Upright Microscope + WI-SNPXLU Revolving Nosepiece + WI-DICTHRA U-LDPXLU20HR DIC Prism + U-KPA Simple Polarizing Intermediate Attachment + U-ANT Analyzer + XLUMPlanFI20XW Objective + WI-DICD Condenser + 32BP775 IR Filter (Polarizer)
Micro-manipulator : Sutter Instrument MPC-385-2 (MPC-200 and ROE-200 and MP-285/M)
Vibration Isolation Platform : Minus k Technology BM-4 Biscuit
Heater Controller : Warner Instruments TC-344B
DAQ Card : National Instrument PCI-6259
BNC Adapter : National Instrument BNC-2090 (using SHC68-68 Cable)
Micro-electrode Amplifier : Axon Instruments MultiClamp 700B
Micro-pipette Headstage : Axon Instruments CV-7B
Water Bath:Precision Scientific 181 STAINLESS
Multichannel Cartridge Pumps: Watson-Marlow* Sci-Q 205U/CA and 205S/CA

Comedi Sample Program(2)

2007-01-09T13:45:00.000-06:00

Source Code 2: tut2.c

#include /* for printf() */
#include

int subdev = 0; /* change this to your input subdevice */
int chan = 0; /* change this to your channel */
int range = 0; /* more on this later */
int aref = 0; /* more on this later */

int main(int argc,char *argv[])
{
comedi_t *cf;
int chan=0;
lsampl_t data;
int maxdata,rangetype;
double volts;

cf=comedi_open("/dev/comedi0");

maxdata=comedi_get_maxdata(cf,subdev,chan);

rangetype=comedi_get_rangetype(cf,subdev,chan);

comedi_data_read(cf,subdev,chan,range,aref,&data); //different from the original comedi demo 2 program here

volts=comedi_to_phys(data,comedi_get_range(cf,subdev,chan,range),maxdata); //different from the original comedi demo 2 program here

printf("%d %g\n",data,volts);

return 0;
}

Compiling:
cc tut2.c -lcomedi -lm -o tut2

Running Result:
32768 0.00015259

Comedi Sample Program(1)

2007-01-07T12:51:00.000-06:00

Source Code:tut1.c

#include /* for printf() */
#include

int subdev = 0; /* change this to your input subdevice */
int chan = 0; /* change this to your channel */
int range = 0; /* more on this later */
int aref = AREF_GROUND; /* more on this later */

int main(int argc,char *argv[])
{
comedi_t *it;
lsampl_t data;

it=comedi_open("/dev/comedi0");

comedi_data_read(it,subdev,chan,range,aref, & data);

printf("%d\n",data);

return 0;
}

Compile:
cc tut1.c -lcomedi -lm -o tut1

Running Result:
32768

Add usb device in VMware

2007-01-06T16:07:00.000-06:00

Prequisite:

1) Make sure your host O/S does not have it mounted or loaded.
2) Make sure /proc/bus/usb is mounted as filesystem type usbfs.
# mount -t usbfs none /proc/bus/usb

Step:
In VMware, Edit the virtual machine settings-> Add a usb controller-> Rebooting the virtual machine-> Enable(load) the usb device in "VM"-"Removable Devices"-"USB deivces"

This will be extremely useful since the MultiClamp is connected through usb port and we can use some portable hard drive with the virtual machine to pass the data to the host machine.

Add sound card driver in VMware

2007-01-06T15:29:00.001-06:00

If your virtual machine does not have a virtual sound adapter in the virtual machine configuration file (.vmx)

With the virtual machine powered off, try adding the following lines to the end of the .vmx file

sound.present = "TRUE"
sound.virtualDev = "es1371"
sound.fileName = "-1"
sound.autodetect = "TRUE"

This requires that have a physical sound card installed in your host.

Linux Script Programing

2007-01-05T16:19:00.001-06:00

How to switch to and from Console mode in GNOME

2007-01-04T17:59:00.000-06:00

To switch to Console mode : Press 'Ctrl + Alt + F1' (F2 - F6)
To switch between consoles in Console mode: Press 'Alt + F1' (F2 - F6)
To switch back to GNOME mode : Press 'Alt + F7'

How to install RTLDC on Ubuntu 6.06 (2)

2007-01-04T15:47:00.000-06:00

The first step to installation is to configure the package, to do this run:

  bash$ make config

Then answer the following questions:

3.2.1.1. Location of Linux Kernel Sources
3.2.1.2. Real-Time Linux Version
3.2.1.3. Location of RTAI/RTLinux sources
3.2.1.4. Comedi Directory
3.2.1.5. Comedilib Directory
3.2.1.6. QT Directory

3.2.1.1. Location of Linux Kernel Sources

You probably won't go wrong by answering /usr/src/`uname -r`/build.

3.2.1.2. Real-Time Linux Version

You should specify either RTAI or RTLinux, depending on which of the packages you installed, if you followed the directions in System Requirements you should have RTAI installed.

3.2.1.3. Location of RTAI/RTLinux sources

If you are using RTAI try /usr/src/rtai or if you are using RTLinux try /usr/src/rtlinux.

3.2.1.4. Comedi Directory

This is the location of the Comedi package. If you followed the instructions in System Requirements then you would answer /opt/comedi.

3.2.1.5. Comedilib Directory

This is the location of the Comedilib package. If you followed the instructions in System Requirements then you would answer /opt/comedilib.

3.2.1.6. QT Directory

This is the location of the QT package. If you followed the instructions in System Requirements then you would answer /usr/lib/qt.

3.2.2. Compilation

Now you are ready to begin compiling so run:

  bash$ make

Sit back and wait for the compilation to finish. If your compilation fails checkout the FAQ or if that doesn't work see Getting Help.

NOTE:

1. You have to make a link
$ ln -s /usr/local/lib/libcomedi-0.7.22.so libcomedi-0.7.22.so

Start FTP Serve on Ubuntu

2007-01-03T14:54:00.000-06:00

1. sudo apt-get install proftpd (stand alone service)
2. sudo cp /etc/proftpd.conf /etc/proftpd.conf_backup
3. sudo gedit /etc/proftpd/proftpd.conf
4. find this section...DenyFilter \*.*/....
5. Add the following line below it
DefaultRoot ~
6. Save the edited file
7. addgroup ftpuser
8. adduser username
9. usermod -G ftpuser username
10. sudo /etc/init.d/proftpd restart

Using latest Comedi download through CVS

2007-01-03T13:29:00.000-06:00

The latest unstable version of Comedi and comedilib can be checked out using anonymous CVS using the commands:

cvs -d :pserver:anonymous@cvs.comedi.org:/cvs/comedi login
cvs -d :pserver:anonymous@cvs.comedi.org:/cvs/comedi co comedi
cvs -d :pserver:anonymous@cvs.comedi.org:/cvs/comedi co comedili

1. cd /comedilib
2. apt-get install bison
3. apt-get install flex
4. apt-get install libtool
5. download automake1.9_for_debian package and install it
(6. if you encounter the calib.lex error, you can "rm lib/calib_*.[ch]")
7. ./autogen.sh
8. ./configure
9. make
10. make install
11. cd /comedi
12. ./autogen.sh
13. ./configure
14. make
15. make install
16. depmod -a
17. make dev

Note: The CVS version of Comedi may not be compatible with rtai_comedi module of current rtai_3.4 version,
which means if you are using RTAI and Comedi, it may not be the "realtime" case and can be very slow.

Linux Networking

2007-01-02T17:26:00.000-06:00